1. Field of the Invention
This invention pertains to the field of targeted modification of cellular DNA in vertebrate cells by homologous pairing using parvoviral vectors, including vectors based on adeno-associated virus (AAV).
2. Background
Previously known methods for introducing defined mutations into mammalian chromosomes by gene targeting involve transfection, electroporation or microinjection (Smithies et al. (1985) Nature 317: 230-234; Thomas et al. (1986) Cell 44: 419-428). These methods, except for microinjection, produce homologous recombination events in only a small fraction of the total cell population, on the order of 10−6 in the case of mouse embryonic stem cells (Doetschman et al. (1987) Nature 330: 576-578; Thomas and Capecchi (1987) Cell 51: 503-512). Thus, the routine use of these methods requires preselection of transformed cells, making it difficult to apply the techniques to normal cells and in vivo applications.
Attempts to use transducing viral vectors to overcome these limitations and achieve chromosomal gene targeting experiments have been performed with retroviral and adenoviral vectors, but the results were not significantly better than can be obtained by transfection, with homologous recombination occurring in 10−5 to 10−6 cells (Ellis and Bernstein (1989) Mol. Cell. Biol. 9: 1621-1627; Wang and Taylor (1993) Mol. Cell. Biol. 13: 918-927).
Adeno-associated virus 2 (AAV) is a 4.7 kb single stranded DNA virus that has been developed as a transducing vector capable of integrating into mammalian chromosomes (Muzyczka (1992) Curr. Top. Microbiol. Immunol. 158: 97-129). Two thirds of integrated wild-type AAV proviruses are found at a specific human chromosome 19 site, 19q13-qter (Kotin et al. (1991) Genomics 10: 831-834; Kotin et al. (1990) Proc. Nat'l. Acad. Sci. USA 87: 2211-2215; Samulski et al. (1991) EMBO J. 10: 3941-3950). The site-specific integration event is a non-homologous recombination reaction that appears to be mediated by the viral Rep protein (Giraud et al. (1995) J. Virol. 69: 6917-6924; Linden et al. (1996) Proc. Nat'l. Acad. Sci. USA 93: 7966-7972). While this feature could prove useful in some applications, AAV vectors with deletions in the viral rep gene have not been found to integrate at this same locus (Russell et al. (1994) Proc. Nat'l. Acad. Sci. USA 91: 8915-8919; Walsh et al. (1992) Proc. Nat'l. Acad. Sci. USA 89: 7257-7261). Southern analysis of integrated rep AAV vector proviruses suggests that integration sites are random (Lebkowski et al. (1988) Mol. Cell. Biol. 8: 3988-3996; McLaughlin et al. (1988) J. Virol. 62: 1963-1973; Russell et al. (1994) supra.; Walsh et al. (1992) supra.) and sequencing of integrated vector junction fragments has confirmed that integration occurs by non-homologous recombination at a variety of chromosomal sites.
Although the development of integrating vectors based on eukaryotic viruses has made possible the efficient introduction of genes into mammalian chromosomes, there are many situations where it would be preferable to modify specific chromosomal sequences. Rather than, for example, introducing a corrected version of gene at a chromosomal location other than the native locus for the gene, one could correct the defective allele at the native locus. This ability could eliminate unwanted chromosomal genotypes and avoid position effects on gene expression. The need for such an ability to modify a preexisting locus is particularly acute in gene therapy, where mutant genes can have dominant effects and tissue-specific controls on expression are often critical.
Thus, a need exists for methods of obtaining specific genetic modification at selected target sites in vertebrate cellular genomes at high frequencies. The present invention fulfills this and other needs.